Alzheimer’s researchers face trial for scientific fraud and defrauding US government

BMJ 2012; 344 doi: (Published 22 May 2012) Cite this as: BMJ 2012;344:e3608
  1. Janice Hopkins Tanne
  1. 1New York

A leading US researcher in Alzheimer’s disease and two Harvard teaching hospitals face trial later this year, charged with using fraudulent scientific data to obtain a grant from the National Institutes of Health. The suit relies on the False Claims Act of 1863, which penalises contractors who defraud the government.

The case is thought to be the first such involving alleged scientific fraud to go to trial in the United States—most previous cases have dealt with contractors defrauding the government by providing faulty supplies.

More lawsuits brought under the False Claims Act on the basis of scientific data manipulation may be expected, said Michael Kohn, of the Kohn, Kohn & Colapinto law firm in Washington, DC, who represents the whistleblower who exposed the fraud. “This is a major breakthrough holding universities accountable for the integrity of reported research results,” he said.

Those charged are Marilyn Albert, a former professor of psychiatry at Harvard Medical School and Massachusetts General Hospital in Boston, who is now director of the division of cognitive neuroscience at Johns Hopkins University School of Medicine, Baltimore, and two Harvard teaching hospitals: Brigham and Women’s Hospital and Massachusetts General Hospital in Boston.

Harvard University and Harvard Medical School were originally named in the suit, but the plaintiffs agreed to dismiss them from the case in 2009, said David Cameron, the medical school’s director of science communications.

The whistleblower is Kenneth James Jones, who was chief statistician for the grant application. He complained that measurements to demonstrate the reliability of the study had been altered.

Whistleblowers who bring cases under the False Claims Act receive financial rewards. If he wins, Jones might receive $5.4m (£3.4m; €4.2m) to $10.8m, as determined by the court.

A lower court had dismissed the Jones’s lawsuit, but he appealed to the First Circuit Court, which decided that the lower court had erred by not considering possible evidence of fraud and said that the case should proceed.

The allegedly fraudulent scientific data involved use of magnetic resonance imaging (MRI) brain scans to predict Alzheimer’s disease. The data were used to support the application for a $15m grant from the National Institute of Aging, part of the National Institutes of Health. The actual grant was about $12.5m.

In the grant application Albert, the principal investigator, theorised that shrinkage of the entorhinal cortex was a predictor of Alzheimer’s disease and would enable doctors to intervene early.

Ron Killiany, a neuroanatomist associated with Boston University School of Medicine, Brigham and Women’s Hospital, and Massachusetts General Hospital, and Teresa Gomez-Isla measured the entorhinal cortex, part of the medial temporal lobe.

Killiany and Gomez-Isla, who were blinded to each participant’s status, manually outlined the entorhinal cortex on MRI scans of 103 people in the study, who were considered normal or control; questionable; or converter (someone who later developed Alzheimer’s disease). Jones carried out statistical analyses of the data to determine whether changes in the volume of the entorhinal cortex could predict which cognitively healthy people would develop Alzheimer’s disease.

According to court documents, Killiany modified his outlining process during 1995 and 1999 and retraced the data of about 30 people before sending the data to the researchers.

In March 2001 Jones saw that there was more than one data set and expressed concern that Killiany’s modifications were responsible for the apparent statistical significance. He asked for a re-examination of Killiany’s changed measurements, which was done. However, Jones was unsatisfied and asked that the scans be remeasured. Albert refused. Soon after, Jones refused to work with MRI data from Killiany’s measurements and lost his job, and “his career came to an end,” said Kohn.

Findings based on the allegedly fraudulent research were published in 2000.1

Kohn told the BMJ that if Killiany’s original measurements had been used, there would have been no statistical significance. “The smallest measures of the control group were selectively enlarged, and this change was solely responsible for providing a scientifically significant result,” he said in an email.

Killiany did not respond to an email from the BMJ asking for comment.

Albert told the BMJ by email, “I am confident there was no misconduct involved. I have no further comment.”

Massachusetts General Hospital and Brigham and Women’s Hospital sent the BMJ identical statements, saying that they were “confident that the researchers acted appropriately and according to the highest standards of scientific integrity. While it is disappointing that additional time and resources will have to be devoted to defending the institution and its investigators, the [institution] remains confident that the resolution of the case will show that the allegations are without worth.”


Cite this as: BMJ 2012;344:e3608


View Abstract

Log in

Log in through your institution


* For online subscription